|Publication number||US4653029 A|
|Application number||US 06/720,794|
|Publication date||Mar 24, 1987|
|Filing date||Apr 8, 1985|
|Priority date||Apr 6, 1984|
|Publication number||06720794, 720794, US 4653029 A, US 4653029A, US-A-4653029, US4653029 A, US4653029A|
|Original Assignee||Hitachi, Ltd.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (6), Non-Patent Citations (2), Referenced by (14), Classifications (11), Legal Events (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention relates to an amplifier consisting of MOSFETs (insulated gate field effect transistors) and to a technique which is effective when used, for example, for a main amplifier for amplifying a read-out signal from a memory array in a dynamic type RAM (random access memory).
The inventor of the present invention developed previously an MOS amplifier such as shown in FIG. 1 which constitutes a main amplifier in a dynamic type RAM. This MOS amplifier consists of two sets of MOS differential amplifiers, and is arranged to receive complementary input signals in mutually opposite phases and to produce complementary output signals from the respective output terminals. A MOS amplifier similar to the arrangement of FIG. 1 is shown in FIG. 6 of U.S. Pat. No. 4,509,147 issued on Apr. 2, 1985, which patent is hereby incorporated by reference.
Although the circuit of FIG. 1 is useful as a main amplifier, the inventor of the present invention determined that it would be desirable to improve its sensitivity. Regarding the structure of the dynamic type RAMs, it is noted that Japanese Laid-Open No. No. 57-082282 is an example of a prior art reference that describes in detail the dynamic type RAM.
An object of the present invention is to provide an MOS amplifier which attains improved sensitivity with a simple circuit construction.
The above and other objects and novel features of the present invention will become more apparent from the following description to be taken in conjunction with the accompanying drawings.
Among the inventions disclosed herein, the following will illustrate a typical example.
MOSFETs Q2 and Q3 (see FIG. 2) connected in differential form are prepared for a pair of amplification MOSFETs Q1 and Q4 which receive input signals and which are connected in differential form. The MOSFETs Q2 and Q3 receive at their drains output currents of current mirror circuits which in turn receive the drain currents of the amplification MOSFETs Q1 and Q4 described above. The gates and drains of these MOSFETs Q2 and Q3 are cross-connected, respectively, in order to utilize positive feedback to improve sensitivity.
FIG. 1 is a circuit diagram showing an example of the main amplifiers developed previously by the inventor of the present application:
FIG. 2 is a circuit diagram of an MOS amplifier in accordance with one embodiment of the present invention;
FIG. 3 is a circuit diagram showing an example of dynamic type RAMs using the main amplifier of the present invention;
FIG. 4 is a specific circuit diagram of a circuit that can be used as the block DOB in FIG. 3; and
FIGS. 5 and 6 are circuit diagrams of the other embodiments of the present invention.
FIG. 2 is a circuit diagram of one embodiment of the present invention. Each circuit element in the drawing is formed on a semiconductor substrate such as single crystal silicon, though not particularly limitative, by a known CMOS production technique (complementary MOS).
Though not particularly limitative, the MOS amplifier of this embodiment can be applied to a main amplifier for amplifying read-out signals appearing on common complementary data lines CDL and CDL in the dynamic type RAM which will be discussed later.
N-channel MOSFETs are formed in a semiconductor substrate made of p-type single crystal silicon, for example, and p-channel MOSFETs are formed in an N-type well region formed on the semiconductor substrate The semiconductor substrate constitutes a common substrate gate of the N-channel MOSFETs, while the N-type well region forms the substrate gate of the P-channel MOSFETs. When connected to the output terminals of a backbias voltage generation circuit, for example, the substrate gate of the N-channel MOSFETs is kept at a suitable negative potential. The substrate gate of the P-channel MOSFETs is kept at the potential of a power terminal Vcc.
The signals of the common complementary data lines CDL and CDL are applied to the gates of amplification MOSFETs Q1 and Q4 that are coupled in differential form. The drains of these amplification MOSFETs Q1 and Q4 are connected to the drains of MOSFETs Q6 and Q8 on the input side of current mirror circuits M1 and M2 that are constituted by P-channel MOSFETs Q6, Q7 and Q8, Q9, respectively. The drains of MOSFETs Q7 and Q9 on the output side of these current mirror circuits M1 and M2 are connected to the drains of N-channel MOSFETs Q2 and Q3, respectively. The sources of these MOSFETs Q2 and Q3 are connected in common to the sources of the amplification MOSFETs Q1 and Q4 in the differential form. To obtain high sensitivity, the gates of MOSFETs Q2 and Q3 are cross-connected to their drains, respectively, thereby forming a positive feedback loop. N-channel MOSFET Q5, its conductance controlled by an operating timing signal φma, is interposed between the common source of the amplification MOSFETs Q1 through Q4 and the ground potential point of the circuit. This MOSFET Q5 forms the operating current of the amplification MOSFETs Q1 through Q4.
The circuit of this embodiment operates in the following manner.
When the timing signal φma is at a low level such as about 0 V, MOSFET Q5 is kept off in response thereto, and the amplification circuit is in the inoperative state. In this inoperative state, the outputs Do and Do are at the high level, that is, at the level of the reset state, and the common source of MOSFETs Q1 through Q4 is at the high level.
When the timing signal φma reaches a high level such as the power source voltage Vcc, MOSFET Q5 becomes conductive in response thereto, and the amplification circuit starts operating.
When conduction of MOSFET Q5 starts, the potential of the common source of MOSFETs Q1 through Q4 drops. One of the differential operation MOSFETs Q1 and Q4 is rendered strongly conductive in accordance with a read-out level that is given in advance to the common complementary data lines CDL and CDL. For instance, if the read-out level of one of the common complementary data line CDL is at a higher level than the other CDL, for example, a greater drain current flows through the amplification MOSFET Q1 than through the amplification MOSFET Q4 when MOSFET Q5 is turned on by the operation timing signal φma. The drain currents of these transistors are applied to the drains of the amplification MOSFETs Q2 and Q3 in the latch form through the current mirror circuits M1 and M2, respectively. In this case, the current applied to the drain of the amplification MOSFET Q2 is greater than that applied to the drain of the amplification MOSFET Q3. Therefore, the drain-source potential of MOSFET Q2, that is, the gate-source potential of MOSFET Q3, is rapidly increased. Therefore, the amplification MOSFET Q3 is controlled so as to conduct a greater current. In this case, the current that is supplied from MOSFET Q9 of the current mirror circuit to the drain of MOSFET Q3 is small, and almost all of this current flows through the amplification MOSFET Q3. This MOSFET Q3 absorbs the current supplied to the gate capacitance of the amplification MOSFET Q2 and attempts to turn off the latter. As a result, a positive feedback operation is executed so that almost all the current supplied from MOSFET Q7 of the current mirror circuit M1 flows into the gate capacitance of the amplification MOSFET Q3. Therefore, MOSFET Q2 is abruptly turned off while MOSFET Q3 is turned on, so that the output signal Do reaches the high level while the output signal Do falls to the low level. These complementary output signals Do and Do are sent out from the external terminal Dout through the output buffer.
When the sources of MOSFETs Q2 and Q3, that are connected in the latch form as shown in FIG. 2, are connected in common to the sources of the differential operation MOSFETs Q1 and Q4, the following desirable circuit operation is executed.
In the transient period of signal amplification, the current mirror circuit M1 produces a current which is proportional to the drain current of MOSFET Q1. Similarly, the current mirror circuit M2 produces a current proportional to the drain current of MOSFET Q4. If MOSFETs Q2 and Q3 connected in the latch form are not yet turned on and off distinctively in the transient period of signal amplification, the output currents of the current mirror circuits M1 and M2 are supplied to the common source of MOSFETs Q1 through Q4 via these MOSFETs Q2 and Q3. These output currents can be regarded as a kind of negative feedback signal.
Here, if the level difference between the common complementary data lines CDL and CDL is small, and if the timing signal φma is generated at a timing at which the level difference given to these data lines CDL and CDL is not increased sufficiently, the difference of the drain current between MOSFETs Q1 and Q4 is small. The operating speed of MOSFETs Q2 and Q3 connected in the latch form is affected by the difference of the drain current of MOSFETs Q1 and Q2. In accordance with this embodiment, the potential of the common source of MOSFETs Q1 through Q4 is raised by the output currents of the current mirror circuits M1 and M2. As a result, the difference of the drain current of MOSFETs Q1 and Q4 is substantially increased.
The potential rise due to the negative feedback operation of the common source of MOSFETs Q1 through Q4 restricts any erroneous operation of MOSFETs Q2 and Q3 connected in the latch form.
Noise which can be regarded as a common mode noise is often applied to the common complementary data lines CDL and CDL through undesirable coupling capacitance such as the gate-source capacitance or gate-drain capacitance of the column switch MOSFETs discussed later.
The negative feedback operation described above reduces the sensitivity of the circuit to the common mode noise without substantially reducing the circuit sensitivity to the signals applied to the data lines CDL and CDL.
In accordance with this embodiment, the output terminals Do and Do are automatically reset or brought to the initial level when the timing signal φma is at the low level, that is, when the amplification circuit is in the inoperative state.
In other words, when the timing signal φma is at the low level and hence MOSFET Q5 is turned off, the potential of the common source of MOSFETs Q1 through Q4 is raised by MOSFET of Q1 through Q4 which is rendered conductive. Here, if the common complementary data line CDL is at a higher level than the data line CDL, the output terminal Do is at the high level while the output terminal Do is at the low level in response thereto. Correspondingly, MOSFET Q2 is turned off and MOSFET Q3 is turned on. The output terminal Do is kept at the high level by a floating capacitance (not shown) depending upon the output terminal Do, even though MOSFET Q5 is turned off.
Therefore, when MOSFET Q5 is turned off, the potential applied to the common source is given to the output terminal Do through MOSFET Q3. As a result, the potential of the output terminal Do is raised. If the common complementary data line CDL is at the high level substantially equal to the power source voltage Vcc, the common source of MOSFETs Q1 through Q4 is raised to about Vcc -Vth1 -Vth2 (where Vth1 and Vth2 are threshold voltages of MSOFETs Q6 and Q1, respectively), and the output terminal Do is likewise raised to about Vcc -Vth1 -Vth2.
In this manner, the setting time of the output terminals Do and Do can be made substantially constant when the amplification circuit is brought again into the operative state, as the levels of the output terminals Do and Do are reset to the initial levels.
In order to bring the initial level of the output terminal Do better coincidence with that of Do, a resetting MOSFET or an equalizer MOSFET may be disposed between the output terminals Do and Do. Alternatively, resetting MOSFETs may be disposed between the output terminals and the power source terminal Vcc. In this case, each of the resetting MOSFETs may be small because MOSFET Q3 or Q2 is operated.
When the complementary output signals are generated by one output signal and the output signal of an inverter circuit receiving the former output signal without using the embodiment shown in FIG. 2, one of the complementary output signals is delayed by the inverter circuit.
In contrast, the complementary output signals obtained from the amplification circuit of this invention are not undesirably delayed, but can be regarded as being changed at substantially the same timing. Therefore, a circuit that receives the complementary output signals can be operated at high speed
FIG. 3 shows a circuit diagram of one example of a dynamic RAM which uses the MOS amplification circuit of this invention as a main amplifier.
In the circuit of this embodiment shown in the drawing, circuit elements constituting a memory array M-ARY consist of IGFETs (Insulated Gate Field Effect Transistors) typified by N-channel MOSFETs.
Two 1-bit memory cells MCs are shown in the drawing, each consisting of a data storage capacitor Cs and an address selection MOSFET Qm. Data stored in the capacitor Cs is expressed as logic "1" or "0" depending on whether or not an electric charge is present in the capacitor Cs.
Data read-out of any memory cell MC is effected by sensing which change occurs in the potential of the data line DL in accordance with a charge quantity stored in the capacitor Cs when the capacitor Cs is connected to the common data line DL by turning on the address selection MOSFET Qm associated with the capacitor Cs of the memory cell being read.
The memory matrix M-ARY has large scale integration and large memory capacity because each memory cell MC is small and many such memory cells are connected to the common data line DL. Moreover, the ratio of the capacity of capacitor Cs to the floating capacitance Co (not shown) of the common data line DL, that is, Cs/Co, is extremely small. Therefore, the potential change of the data line DL due to the charge quantity stored in the capacitor Cs is an extremely sensitive signal.
A dummy cell DC is disposed as a reference for detecting such a sensitive signal. The dummy cell DC is fabricated in the same production conditions and with the same design constants as the memory cell Mc except that the capacitance of its capacitor Cd is substantially half that of the capacitor Cs of the memory cell MC. The capacitor Cd is charged to the ground potential by MOSFET Qd' before addressing.
Since the capacitance of the capacitor Cd is set to about half of that of the capacitor Cs as described above, it can form a reference voltage which is substantially equal to half of the readout signal from the memory cell MC.
In the drawing, symbol SA represents a sense amplifier which expands the difference of such a potential change caused by addressing described above to a sense period that is determined by timing signals (sense amplifier control signals) φpa1 and φpa2. (Its operation will be described elsewhere.) Its input-cutput nodes are connected to a pair of complementary data lines DL and DL that are disposed in parallel with each other. The numbers of memory cells to be connected to the complementary data lines DL and DL are equal to each other so as to improve detection accuracy, and one each dummy cell is connected to each of DL and DL. Each memory cell is folded between one word line WL and one of the pair of the complementary data lines. In the case of a memory matrix of a folded bit line construction, each word line WL crosses both of the data line pairs. Therefore, even when a noise component generated in the word line WL is carried out to the data line due to electrostatic coupling, the noise component appears equally on both of the data line pairs DL and DL. Such a common mode noise is neglected by the differential type sense amplifier SA.
When the memory cell MC connected to one of the complementary data line pair DL and DL is selected in addressing described above, one of the dummy word line pair DWL and DWL is always selected so that the dummy cell DC is connected to the other of the data lines.
The sense amplifier SA has a pair of MOSFETs Q11 and Q12 and that are cross-connected with each other, and the positive feedback operation of these transistors differentially amplifies the delicate signal that appears on the complementary data lines DL and DL. This positive feedback operation is dividedly carried out in two steps. The positive feedback operation of the first step is started simultaneously with the start of conduction of MOSFET Q17 which has relatively small conductance and which is rendered conductive by a relatively rapid timing signal φpa1. A higher data line potential, which is set on the basis of the potential difference given to the complementary data lines DL and DL by addressing, drops at a lower speed while a lower data line potential drops at a higher speed, thereby expanding the potential difference between them.
The circuit operation of the second step is executed at the time at which the potential difference described above becomes great to certain extents. That is, MOSFET Q18 which has relatively great conductance is rendered conductive by the timing signal φpa2, and the lower data line potential described above rapidly drops. Since the operation of the sense amplifier SA is dividedly carried out in two steps in the manner described above, the drop of the higher potential can be prevented. When the lower potential drops below the threshold voltage of the cross-connected MOSFETs, the positive feedback operation is finished, the drop of the higher potential remains at a level lower than the power source voltage Vcc but higher than the threshold voltage, and the lower potential finally reaches the ground potential (0 V).
When addressing described above is effected, the stored data that was about to be destroyed is recovered since it receives a high or low level potential such as that which is obtained by the sensing operation. However, if the high level drops by a predetermined level with respect to the power source voltage Vcc as described already, the logic is erroneously read as "0" in the course of repetition of read-out and re-writing operations. An active restore circuit AR is disposed in order to prevent this error. This active restore circuit AR does not affect the low level signal at all but selectively boosts only the high level signal to the power source voltage Vcc.
The data line pair DL and DL shown as being representative in the drawing is connected to the common complementary data line pair CDL and CDL via MOSFETs Q13 and Q14 that constitute a column switch CW. The data line pair shown as another representative example are also connected to the complementary data line pair CDL and CDL via similar MOSFETs Q15 and Q16. This common complementary data line pair CDL and CDL is connected to the input terminal of a data output buffer DOB including an output amplifier and to the output terminal of a data input buffer DIB. The data output buffer DOB consists of the main amplifier which receives the signals of the common complementary data lines CDL and CDL and is shown in FIG. 2, and an output buffer which delivers the signal from the external terminals of the main amplifier.
A row-and-column decoder RC-DCR receives an internal complementary address signal generated by the address buffer ADB, and generates a selection signal for one word line, dummy word line and column switch. Thus, addressing to the memory cells and dummy cells is effected. In other words, external address signals AXo -AXi are taken into the address buffer ADB in synchronism with the timing signal φar that is generated by a row address strobe signal RAS. The internal address signal produced from the address buffer ADB is transmitted to the decoder RC-DCR. Next, the selecting operation of a predetermined word line and dummy word line is effected by a word line selection timing signal φx. External address signals AYo -AYi are taken into the address buffer ADB in synchronism with the timing signal φac generated by the column address strobe signal CAS, and the internal address signal formed thereby is transmitted to the decoder RC-DCR. When the data line selection timing signal φy is generated, the selection operation of the data lines is effected.
A timing control circuit TG receives the address strobe signals RAS and CAS supplied from outside and a write enable signal WE, and generates various timing signals besides the timing signals described above as representative.
Though not particularly limitative, the peripheral circuit of the column selection system in the dynamic type RAM of this embodiment consists of a CMOS static type circuit. This makes it possible to read out a plurality of continuous bits by changing the column address signals under the state in which one word line is selected.
FIG. 4 shows a specific circuit diagram of the data output buffer circuit DOB that is shown in FIG. 3. The data output buffer circuit DOB consists of the main amplifier MA having the same construction as the amplifier shown in FIG. 2 and the output buffer OB. The output buffer OB consists of CMOS NAND circuits G1, G2, CMOS inverter circuits IV1, IV2 and N-channel output MOSFETs Q10, Q11. Thus, the output buffer OB constitutes a tri-state circuit.
When the timing signal φop is at the low level, the outputs of the inverter circuits IV1 and IV2 drop to the low level in response thereto, so that the output MOSFETs Q10 and Q11 are turned off. Therefore, the output of the data output buffer DOB is brought into the high impedance state.
When the timing signal φop is at the high level, one of the outputs of the inverter circuits IV1 and IV2 is at the high level with the other being at the low level in accordance with the complementary output signals supplied from the main amplifier. In consequence, the output MOSFETs Q10 and Q11 are turned on complementarily, and the external terminal Dout is brought to the high or low level.
In accordance with this embodiment, the output MOSFETs Q10 and Q11 can be made of the single channel type, that is, of the N-channel type, although the dynamic RAM is fabricated as a whole by CMOS integration circuit technique. This construction provides the following advantages.
The output MOSFETs Q10 and Q11 must have sufficiently low ON resistance in order to sufficiently drive a relatively heavy capacitance load that is connected to the external terminal but is not shown in the drawing. Therefore, these transistors are large. If MOSFET Q10 consists of a P-channel MOSFET, it must be two or three times as large as an N-channel MOSFET Q11. As a result, the area occupied by MOSFETs Q10 and Q11 is large. If MOSFET Q10 consists of an N-channel transistor in the same way as MOSFET Q11, however, the area they occupy can be reduced.
If an abnormal voltage of an undesirable level is applied to the external terminal in the CMOS structure, a parasitic thyristor operation will occur, and the power source wiring layer on the semiconductor substrate will be damaged by abnormal power source current due to the parasitic thyristor operation.
If both MOSFETs Q10 and Q11 are of the single channel type MOSFETs, however, parasitic thyristor operation does not substantially occur even if an abnormal voltage is applied to the external terminal Dout.
FIG. 5 shows a circuit diagram of a data output buffer DOB in accordance with another embodiment of the invention. In this embodiment, the main amplifier MA consists of cascaded amplifiers MA1 and MA2 in order to obtain higher sensitivity. Each of the amplifiers MA1 and MA2 is the same as the amplifier shown in FIG. 2.
Though not particularly limitative, the amplifier MA1 is actuated by the timing signal φma1, while the amplifier MA2 is actuated by the timing signal φma2 which is somewhat delayed from the timing signal φmal.
Since each of the amplifiers MA1 and MA2 has the complementary input terminal and the complementary output terminal, or since each amplifier MA1, MA2 consists of a full differential amplifier, the main amplifier MA can be made to be responsive to the input signals without being made to be responsive substantially to the noise that is unavoidably applied to the common complementary data lines CDL and CDL and to the output terminals Dol and Dol via an undesirable coupling capacitance.
FIG. 6 is a circuit diagram of the amplifier in accordance with still another embodiment of the present invention. In the amplifier of this embodiment, P-channel MOSFETs Q6 and Q8 as the load to the differential N-channel MOSFETs Q1 and Q4 are connected in the latch form. P-channel MOSFETs Q12 and Q13 for resetting are connected in parallel with MOSFETs Q6 and Q8, respectively.
When the timing signal φma is at the low level, the amplifier is inoperative. In this case, resetting MOSFETs Q12 and Q13 are turned on, and the output terminals Do and Do are reset almost to the level of the power source voltage Vcc.
When the timing signal φma is at the high level, MOSFET Q5 is turned on while MOSFETs Q12 and Q13 are turned off. As a result, the amplifier becomes operative.
The amplifier of this embodiment has a reduced number of circuit elements.
The present invention described with regard to the various embodiments above has the following effects and advantages.
(1) The amplifier can obtain a high gain by means of the active load circuit consisting of the current mirror circuits. In the two MOS differential amplification circuits, MOSFETs that receive the differential currents are connected in the latch form. Therefore, the high gain described above and the positive feedback operation operate synergistically, and a high sensitivity MOS amplifier can be obtained.
(2) Since only a part of the wiring of a differential amplifier having two active load circuits is changed relative to the circuit of FIG. 1, a high sensitivity MOS amplifier can be constituted without increasing the number of circuits elements.
(3) Since the amplifier has high sensitivity, the common complementary data lines can be responsive even to a delicate signal when the amplifier is utilized as the main amplifier in a dynamic RAM. Therefore, the operation timing of the main amplifier can be sped up, and read-out at a higher speed can be accomplished.
(4) Due to the effect (3) described above, the operation margin of the main amplifier can be expanded by its high sensitivity.
(5) When the amplifier of the invention is applied to the main amplifier of a dynamic RAM having the function of continuously reading out data of a plurality of bits by means of the column change-over by fabricating the column system selection circuit in the form of a static type circuit, continuous read-out of the data of a plurality of bits can be made at a high speed because of the effect (3) described above.
Although the invention has thus been described with reference to some preferred forms thereof, the invention is not particularly limited thereto, but can, of course, be modified in various ways without departing from the scope and spirit thereof. For example, N-channel MOSFETs and P-channel MOSFETs may be replaced by one another, respectively. In the amplification MOSFETs Q1 and Q4, the signal to be amplified may be applied to the gate of one of these transistors with a reference voltage being applied to the gate of the other.
Although the description given above deals with the case where the amplifier of the invention is applied to the main amplifier of the dynamic RAM which is the background and technical field of the invention, it is not particularly limited thereto, but can be widely utilized as an MOS amplifier.
It is to be understood that the above-described arrangements are simply illustrative of the application of the principles of this invention. Numerous other arrangements may be readily devised by those skilled in the art which embody the principles of the invention and fall within its spirit and scope.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4136292 *||Aug 25, 1977||Jan 23, 1979||Tokyo Shibaura Electric Co., Ltd.||Voltage sensing circuit of differential input type|
|US4247791 *||Apr 3, 1978||Jan 27, 1981||Rockwell International Corporation||CMOS Memory sense amplifier|
|US4479202 *||May 11, 1983||Oct 23, 1984||Tokyo Shibaura Denki Kabushiki Kaisha||CMOS Sense amplifier|
|US4509147 *||Jun 1, 1982||Apr 2, 1985||Hitachi, Ltd.||High speed semiconductor memory device having a high gain sense amplifier|
|US4521704 *||Jan 31, 1983||Jun 4, 1985||Mitsubishi Denki Kabushiki Kaisha||Differential amplifying circuit|
|JPH10291A *||Title not available|
|1||Balasubramanian et al., "Two Device per Bit, Precharged ROS with Differential Sensing", IBM Technical Disclosure Bulletin, vol. 19, No. 1, Jun. 1976, pp. 164-165.|
|2||*||Balasubramanian et al., Two Device per Bit, Precharged ROS with Differential Sensing , IBM Technical Disclosure Bulletin, vol. 19, No. 1, Jun. 1976, pp. 164 165.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US4871933 *||Aug 31, 1988||Oct 3, 1989||Actel Corporation||High-speed static differential sense amplifier|
|US4871978 *||Aug 10, 1988||Oct 3, 1989||Actel Corporation||High-speed static differential sense amplifier|
|US4984206 *||Jun 27, 1989||Jan 8, 1991||Mitsubishi Denki Kabushiki Kaisha||Random access memory with reduced access time in reading operation and operating method thereof|
|US5014245 *||Jun 20, 1990||May 7, 1991||Kabushiki Kaisha Toshiba||Dynamic random access memory and method for writing data thereto|
|US5214317 *||May 4, 1992||May 25, 1993||National Semiconductor Corporation||CMOS to ECL translator with incorporated latch|
|US5248946 *||Dec 11, 1991||Sep 28, 1993||Mitsubishi Denki Kabushiki Kaisha||Symmetrical differential amplifier circuit|
|US5251175 *||Mar 11, 1992||Oct 5, 1993||Fujitsu Limited||Semiconductor memory device|
|US5276643 *||Nov 26, 1991||Jan 4, 1994||Siemens Aktiengesellschaft||Integrated semiconductor circuit|
|US5880582 *||Sep 3, 1997||Mar 9, 1999||Sumitomo Electric Industries, Ltd.||Current mirror circuit and reference voltage generating and light emitting element driving circuits using the same|
|US5909049 *||Feb 11, 1997||Jun 1, 1999||Actel Corporation||Antifuse programmed PROM cell|
|US6181169 *||Oct 28, 1999||Jan 30, 2001||Integrated Technology Express, Inc.||High-speed rail-to-rail comparator|
|US6281821 *||Sep 30, 1997||Aug 28, 2001||Jason P. Rhode||Digital-to-analog converter with power up/down transient suppression and automatic rate switching|
|US20140226425 *||Apr 22, 2014||Aug 14, 2014||Micron Technology, Inc.||Current mode sense amplifier with passive load|
|EP1096677A2 *||Oct 24, 2000||May 2, 2001||Nec Corporation||Latch circuit with a small number of nodes for high speed operation|
|U.S. Classification||365/205, 330/257, 327/54, 365/208, 327/53|
|International Classification||H03K3/356, G11C11/409|
|Cooperative Classification||H03K3/35613, H03K3/356034|
|European Classification||H03K3/356D2, H03K3/356G2D|
|Oct 15, 1986||AS||Assignment|
Owner name: HITACHI, LTD., TOKYO 100, JAPAN
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNOR:SATO, KATSUYUKI;REEL/FRAME:004613/0985
Effective date: 19850528
|Jul 2, 1990||FPAY||Fee payment|
Year of fee payment: 4
|Jul 1, 1994||FPAY||Fee payment|
Year of fee payment: 8
|Oct 13, 1998||REMI||Maintenance fee reminder mailed|
|Mar 21, 1999||LAPS||Lapse for failure to pay maintenance fees|
|Jun 1, 1999||FP||Expired due to failure to pay maintenance fee|
Effective date: 19990324